CtrlK
BlogDocsLog inGet started
Tessl Logo

gastown

Multi-agent orchestrator for Claude Code. Use when user mentions gastown, gas town, gt commands, bd commands, convoys, polecats, crew, rigs, slinging work, multi-agent coordination, beads, hooks, molecules, workflows, the witness, the mayor, the refinery, the deacon, dogs, escalation, or wants to run multiple AI agents on projects simultaneously. Handles installation, workspace setup, work tracking, agent lifecycle, crash recovery, and all gt/bd CLI operations.

62

Quality

73%

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Advisory

Suggest reviewing before use

Fix and improve this skill with Tessl

tessl review fix ./skills/tools/gastown/skills/gastown/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

47%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This skill has strong workflow clarity with excellent validation checklists and error recovery guidance, but is severely undermined by verbosity. Roughly half the content is persona coaching, creative freedom encouragement, and interaction style guidance that Claude doesn't need spelled out at this length. The command reference and system readiness sections are the strongest parts, while the repeated 'CALL THE TOOL' emphasis and extensive example narratives bloat the file significantly.

Suggestions

Cut the 'Creative Freedom' section entirely (Claude doesn't need permission to make ASCII art or be proactive) and reduce 'Interaction Style' to just the AskUserQuestion JSON schema with one example - save ~80 lines.

Move the lengthy examples section (Learning, Setup, Operating examples) to a separate references/examples.md file, keeping only one compact example inline.

Remove the 'Your Identity' section - Claude doesn't need to be told it's an expert or that it 'never guesses'; the verification checklist already handles this operationally.

Consolidate the repeated 'CALL the AskUserQuestion tool' instructions into a single clear directive instead of restating it 5+ times across different sections.

DimensionReasoningScore

Conciseness

Extremely verbose at 500+ lines. Massive amounts of content Claude already knows how to do (interaction patterns, persona instructions, creative freedom examples, ASCII art examples). The 'Creative Freedom' section alone is unnecessary padding - Claude doesn't need to be told it can make ASCII art or be proactive. The capability checklist at the end, the extensive examples section, and repeated emphasis on 'CALL the tool' (stated 5+ times) all waste tokens.

1 / 3

Actionability

Commands are listed with clear syntax and the workflow diagrams show concrete steps. However, the actual executable guidance is mixed with extensive persona/style instructions. The commands section is useful but lacks complete examples with real output. Many 'examples' are illustrative narratives rather than executable sequences - e.g., the setup example says '*installs gt and bd*' rather than showing actual install commands.

2 / 3

Workflow Clarity

The System Readiness Checklist is excellent with explicit validation checkpoints, blocker identification, error severity guide, and a full flow verification sequence (create bead → sling → peek → witness → refinery → git log). The 'Never Assume - Verify Everything' section adds proper feedback loops. The error severity table with actions is well-structured.

3 / 3

Progressive Disclosure

References to external files (tutorial.md, setup.md, commands.md, concepts.md, troubleshooting.md) are well-organized with a clear table and grep navigation strategies. However, no bundle files were provided to verify these references exist. The SKILL.md itself is monolithic - much of the persona, creative freedom, interaction style, and examples content could be in separate reference files, keeping the main skill leaner.

2 / 3

Total

8

/

12

Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

This is a strong skill description that excels across all dimensions. It provides a clear functional summary, an explicit 'Use when' clause with extensive and distinctive trigger terms, and lists concrete capabilities. The unique domain-specific vocabulary makes it virtually impossible to confuse with other skills.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions: installation, workspace setup, work tracking, agent lifecycle, crash recovery, and gt/bd CLI operations. These are clear, actionable capabilities.

3 / 3

Completeness

Clearly answers both 'what' (multi-agent orchestrator handling installation, workspace setup, work tracking, agent lifecycle, crash recovery, CLI operations) and 'when' (explicit 'Use when' clause with extensive trigger terms and scenarios).

3 / 3

Trigger Term Quality

Extensive coverage of natural trigger terms including the tool name ('gastown', 'gas town'), CLI commands ('gt commands', 'bd commands'), domain-specific concepts ('convoys', 'polecats', 'crew', 'rigs', 'beads', 'hooks', 'molecules'), role names ('the witness', 'the mayor', 'the refinery', 'the deacon'), and general concepts ('multi-agent coordination', 'multiple AI agents'). Users would naturally mention these terms.

3 / 3

Distinctiveness Conflict Risk

Highly distinctive with unique domain-specific terminology ('gastown', 'convoys', 'polecats', 'the witness', 'the mayor', 'gt/bd CLI'). These terms are extremely unlikely to conflict with any other skill.

3 / 3

Total

12

/

12

Passed

Validation

90%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation10 / 11 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (705 lines); consider splitting into references/ and linking

Warning

Total

10

/

11

Passed

Repository
numman-ali/n-skills
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.